Conservation Easement Tax Credits

Abstract

Public concern about urban sprawl has spurred states to offer incentives for land preservation, such as state income tax credits granted to donors of conservation easements. This study examines the effect of these tax credits on conservation easement donations to regional land trust organizations in the lower forty-eight states during the period from 1994 to 2005. Regression results indicate that a state income tax credit has a statistically significant impact on conservation easement donations only during the initial period of implementation. The results also show that donation rates are significantly impacted by the extent of state park land holdings.

Keywords

conservation easement land trust tax policy environmental policy

Introduction

In the past several years, the legislatures of several states have enacted laws providing an income tax credit for taxpayers who agree to limit future development of their property by donating a conservation easement to a land trust (Conservation Resource Center 2007). These tax incentives are a manifestation of the growing support for public policies to encourage open space preservation.

Open space land may provide benefits to the surrounding community that are lost when land is developed (Berry 1976; Coughlin, Berry, and Plaut 1978; Anderson 1993; Kline and Wichelns 1998; Brueckner 2000; Irwin 2002). If owners of open space land are not compensated for the social benefits of leaving open space undeveloped, more land will be developed than is socially desirable.

There is evidence that open space provides benefits to the community. Berry (1976) suggests that undeveloped land has contemplative value in that people derive pleasure from the mere knowledge that certain areas are protected from development. Open space also provides aesthetic benefits in that many people enjoy viewing natural landscapes and escaping from heavily developed urban areas (Berry 1976; Coughlin, Berry, and Plaut 1978; Anderson 1993; Kline and Wichelns 1998; Brueckner 2000). Nearby property owners often reap a portion of these external benefits of open space in the form of higher values for their own property (Irwin 2002). Furthermore, open space preservation often improves environmental quality. Hellerstein et al. (2002) describe some of the environmental benefits of open space preservation, such as wildlife habitat preservation, soil conservation, and watershed protection.

The purpose of this research is to examine if these state income tax credits increase conservation easement donation rates. By granting these tax credits, states are essentially purchasing open space preservation with tax expenditures. The magnitude of their effect is an important factor in deciding whether such tax credits are an effective use of public funds.

Incentives for the Preservation of Open Space

Defenders of Wildlife (2006) describes a wide variety of incentives that states grant to property owners who forgo land development and preserve open space. Some subsidies involve direct payments to property owners for taking specific actions, such as preserving habitat for endangered species. Other subsidies are more general, such as income tax credits for protecting land from development by means of a conservation easement. These subsidies are intended to increase the amount of open space preserved by lowering the opportunity cost of withholding land from development.

A survey by Defenders of Wildlife (2002) found that over forty states provide some type of tax incentive to property owners who support conservation goals, such as the preservation of open space. The most common tax incentive is current use assessment, a property tax incentive that reduces property taxes on undeveloped land such as farmland or timberland. Property tax incentives encouraging the preservation of agricultural land are the most prevalent, existing in all fifty states (Defenders of Wildlife 2006). Less common, but growing in popularity, are state income tax credits for donations of conservation easements to land trusts.

A conservation easement preserves ecological characteristics of property by limiting the owner’s rights to develop it. For example, a conservation easement might limit development of a property parcel to preserve open space, limit damage to wetlands, or to protect endangered species habitat (Boyd, Caballero, and Simpson 1999; Parker 2002). A conservation easement also grants affirmative rights that allow the easement holder to monitor and enforce compliance by the property owner with the terms of the easement.

Land trusts are nonprofit organizations that work to conserve land, in part, by purchasing or accepting donations of conservation easements and monitoring compliance by the land owners (Land Trust Alliance [LTA] n.d.). The first land trust was formed in Massachusetts in 1891 (Brewer 2003). As of December 31, 2005, there were 1,667 state and local land trusts holding conservation easements protecting approximately 6.2 million acres (LTA 2006).

The Impact of Preservation Tax Incentives

There has been very little empirical research examining the effects of conservation easement tax credits on easement donations. A report by the Conservation Resource Center (2007) compared conservation easement donations in particular states before and after the enactment of a tax credit and concluded that the credits lead to an increase in donations. Their comparisons, however, did not account for factors other than the tax credits that might have influenced easement donations. Other studies have addressed other types of tax incentives for open space preservation, suggesting the factors that might influence the effect of tax credits on conservation easement donations.

Simpson (2002) presents a theoretical model of the effects of property tax and income tax incentives on preservation of open space land by developers. In the case of property tax incentives, Simpson predicts that larger property tax incentives for open space preservation will increase the amount of open space withheld from development. In the case of income tax incentives, he predicts that developers faced with higher income tax rates will retain more open space land. However, he also predicts that the effect of the income tax incentive will decrease as the value of developed land increases relative to that of open space land.

Coughlin, Berry, and Plaut (1978) review two studies on the effect of differential property tax assessment on the conversion of Ohio farmland to nonagricultural uses. They find that in counties with highly productive farmland, the tax incentive has no significant effect on the rate of development of farmland. They find that in areas where the farmland is not very productive, tax incentives slow the loss of farmland to development by allowing marginally productive farmland to stay in agricultural use. Finally, they find that in highly urbanized counties, where there is great demand for land by developers, the property tax savings are very small compared to the opportunity costs of leaving the land undeveloped and the incentives are ineffective.

Hellerstein et al. (2002) examine a program compensating farmers who withhold farmland from development. They find that more land is preserved in counties where mean income, population pressure,¹ agricultural land density, rates of population pressure growth, and rates of agricultural land density loss are higher.

Finally, Daugherty (1977) analyzes the federal tax savings generated by donating a conservation easement to a land trust organization. He calculates the tax savings for taxpayers in a wide variety of situations and determines that donations make financial sense only for taxpayers with high incomes or large amounts of taxable wealth in their estates.

This review of literature highlights the importance of controlling for demographic variables and other factors when studying the effects of preservation tax incentives. The next section describes the methods of analysis used in this study to estimate the effect of state income tax credits on the donation of conservation easements to land trust organizations.

Research Design

This analysis is based on a theory of conservation easement donation in which conservation-minded land owners weigh the costs and benefits of donating a conservation easement to a land trust when making donation decisions. When the costs of donation are relatively great, land owners will be less likely to donate or will tend to make smaller donations. Conversely, when the perceived benefits of donation are relatively great, they will be more likely to donate or to make larger donations.

The opportunity cost of foregone future development profits is a major cost of donating a conservation easement. Tax incentives reduce this opportunity cost by subsidizing donations. State conservation easement tax credits lower this cost by reducing a donor’s state income tax liability. Even in the absence of a specific tax credit, donation decisions may be influenced by tax policy. The federal tax code and most state tax codes allow a deduction for charitable contributions. The availability of this deduction reduces the after-tax cost of a conservation easement donation, especially for high-income donors.

Measuring Conservation Easement Donations

The Land Trust Alliance (LTA), an advocacy organization that promotes voluntary private land conservation and represents state and local land trusts, conducts periodic censuses to determine the number of land trusts operating in each state and the amount of land they protect through conservation easements or direct ownership. The LTA census data include statewide land trust holdings as of December 31, 1994, 1998, 2000, 2003, and 2005. These census data are used to calculate the net increase in conservation easement acreage held by land trusts in each of the lower forty-eight states during four consecutive periods: a four-year period beginning January 1, 1995; a two-year period beginning January 1, 1999; a three-year period beginning January 1, 2001; and a two-year period beginning January 1, 2004.

The dependent variable in this analysis is defined as the net conservation easement donation rate for each state, expressed as acres donated per 1,000 acres of land eligible for donation, divided by the number of years in the period. Land eligible for donation is defined as the total land area in the state; minus developed nonfederal land, which has little conservation value and is unlikely to be eligible for donation; minus land owned by the federal government or contained within state park systems, which is unlikely to be encumbered by a conservation easement; minus land already protected by a state or local land trust.² It is necessary to divide by the number of years in each period because the periods are of different lengths.

A close examination of the data reveals that there are a small number of extreme outliers in the independent variable. In 95 percent of the observations, states experienced donation rates no higher than 3 acres per 1,000 acres eligible per year. In all but three observations, states experienced donation rates below 5 acres per 1,000 acres eligible per year. In these three outlying observations, the states of Vermont and Maine, neither of which provide a tax credit for conservation easement donations, experienced donation rates far exceeding not only the rates of other states but also their own donation rates during prior or succeeding periods.

During the 1999–2000 period, Vermont experienced a donation rate of 17.6 acres per 1,000 acres eligible per year, a rate more than five times the rate it experienced in any other period. Maine experienced a donation rate during the 2001–03 period of 17.9 acres per 1,000 acres eligible per year and a rate of 9.8 in the following period. Maine’s easement donation rate was never greater than 0.4 acres per 1,000 acres eligible per year during any other period.

A review of contemporaneous news sources indicates that these surges in conservation easement donations in Maine and Vermont may have been stimulated by the efforts of strong coalitions of conservation supporters along with the availability of a large quantity of environmentally sensitive land in the Northern Forest, stretching across Vermont, Maine, New Hampshire, and New York (Alcoba 2003). For example, in 1999, a coalition of conservation supporters including the State of Vermont, The Conservation Fund, and numerous other private foundations and individual donors arranged for the purchase of 106,000 acres of forest land from the Champion International Corporation for $20 million. While portions of the purchased land were then transferred to state and federal ownership, a major portion—84,000 acres—was transferred to a private forestry company after being protected by conservation and public access easements (Business Wire 1999).

In Maine, too, multi-organization partnerships resulted in the protection of extremely large tracts of forest land by means of conservation easements. In 2001, the State of Maine, the New England Forestry Foundation (NEFF), other private foundations, and individual donors collaborated to purchase a single conservation easement protecting more than 762,000 acres of forest land from development (NEFF 2001).³ In 2005, a $7 million grant from Wal-Mart allowed The Conservation Fund and the NEFF to purchase a conservation easement protecting 312,000 acres.

There are also some unusual outliers at the lower end of the distribution of donation rates, with sixteen observations of negative conservation easement donation rates. At first glance, a negative donation rate would seem to imply that acreage that was previously protected by conservation easements was returned to unprotected status. That is unlikely, however, because conservation easements are perpetual. Rob Aldrich, Director of Communications for the LTA, explained that some of the negative donation rate observations were caused by errors that resulted in double counting of donation acreage. These data entry errors were not uncovered until after the land trust census report had gone to press. When the errors were corrected in the following census, it appeared as if easement acreage had decreased (R. Aldrich, Personal communication, September 23, 2010).

Negative observations, however, are not necessarily erroneous. Land trusts conserve land by several methods, including outright ownership, so a net decrease in easement acreage can also be an indication that land trusts had taken possession of land that had previously been protected by an easement. In other cases, land trusts may hold property for a short time and then convey the property to another organization. Any of these transfers may result in a reduction of conservation easement acreage in a state (R. Aldrich, Personal communication, September 23, 2010).

These extreme and unusual outliers have the potential to confound the analysis and prevent the accurate estimate of tax credit effects. First, ordinary least squares (OLS) regression and panel regression, both of which are used to analyze these data, are sensitive to the presence of outliers in the data. Consequently, the unusually high donation rates resulting from large donations that are not stimulated by state tax credits may interfere with the accurate estimation of the effects of tax credits on donation rates. Similarly, the apparently negative donation rates with the largest magnitudes are the most likely to represent erroneous data resulting from data entry errors which can confound the analysis.

Furthermore, the theory of conservation easement donation underlying this analysis is predicated on the assumption that conservation easement donations occur at a relatively low and steady rate, influenced by landowner values, tax considerations, development pressures, and the availability of suitable land for preservation.⁴ This theory does not account for infrequent and extremely large acquisitions like those that occurred in Maine and Vermont.

To avoid the confounding effects of these outliers, a limited data set is used in the analysis. The limited data set excludes observations with easement donation rates above the 95th and below the 5th percentiles. This trimming rule is used, in part, because it is balanced, removing extreme outliers at both ends of the distribution. The 5th and 95th percentiles were chosen as the trimming boundaries in order to preserve as much data as possible. Trimming at the 1st and 99th percentiles would not exclude all of the most problematic high-end outliers. Trimming at the 10th and 90th percentiles would exclude observations that cannot legitimately be called outliers.

Hypotheses and Independent Variables

The theory of conservation easement donation underlying this analysis predicts that conservation easement donation rates will vary depending on the availability of a tax credit, other characteristics of each state’s tax system, economic and demographic conditions, and citizen attitudes toward conservation. The hypotheses derived from this theory are listed below, with each followed by a brief explanation and a description of variables used in the analysis.

Hypothesis 1: Conservation easement donation rates will be greater in states that provide donors with tax credits than in states that provide no tax credits.

Hypothesis 2: Conservation easement donation rates will be greater in states that provide more generous tax credits than in states that provide no or less generous credits.

Simpson’s (2002) analysis predicts that tax subsidies will tend to increase donation rates and that more generous subsidies will have greater impacts. The presence of a tax credit is indicated using a variable that takes a value of zero if a state did not provide a tax credit at any time during the period and a value of one if a state provided a tax credit during the entire period. If a state implemented a tax credit partway through a period, then the variable takes a value representing the portion of the period during which the tax credit was in effect.⁵

The relative generosity of the tax credits is accounted for by three variables. The first variable is an alternative to the simple indicator variable described above. This variable is equal to the percentage of the easement’s fair market value that is used to calculate the tax credit, multiplied by the proportion of the period during which the tax credit was in effect.⁶

The other two variables are simple indicator variables. One variable indicates whether a state’s tax credit is transferrable, allowing a low-income donor with insufficient tax liability to use the complete tax credit to sell the tax credit to a taxpayer with greater tax liability. The other variable indicates whether the value of the tax credit is capped at $100,000 or less. Land trust experts believe that capping a tax credit at or below $100,000 severely limits its effectiveness (Conservation Resource Center 2007).

Hypothesis 3: States will experience a reduction in donations during the period immediately prior to implementation of a tax credit.

Hypothesis 4: Tax credits will have a larger impact during their initial period of availability than during subsequent periods.

In periods prior to enactment, potential donors are likely to be aware that the legislature is considering enactment of a tax credit and delay making donations until the tax credit becomes available. Consequently, there may be a decline in conservation easement donations during the period just before implementation of a tax credit and a surge in donations during the initial period of implementation. To account for this potential temporal shift, two additional variables are included in the analysis. One variable takes a value of one during the period before a tax credit becomes effective and zero otherwise. The other variable takes a value equal to the variable measuring the tax credit during the first period in which a tax credit is in effect and zero otherwise.

Hypothesis 5: Conservation easement donation rates will be greater in states in which taxpayers face high marginal income tax rates than in states with low marginal income tax rates.

Even in states with no tax credit, taxpayers are eligible to take a charitable deduction for a conservation easement donation when calculating federal income taxes. The federal deduction reduces taxable income and, in most states with an income tax, the taxpayer’s state income tax liability. Simpson (2002) predicts that higher tax rates will tend to stimulate more donations because of the tax savings resulting from donations.

Tax rates are measured with a variable that represents the combined state and federal marginal income tax rate on high-income tax payers.⁷ This variable is determined by computing the additional state and federal income tax liability on $1,000 of additional income earned by hypothetical married, jointly filing taxpayers with an initial income of $1,500,000 (see Feenberg and Coutts 1993; National Bureau of Economic Research n.d.).

Hypothesis 6: Conservation easement donation rates will be greater in states in which taxpayers face high levels of property taxation than in states with low property taxes.

High property taxes increase the costs of holding land in an undeveloped state and may, therefore, encourage owners to develop the property in order to generate revenue from it. If a landowner has no plans to develop the property, then he or she can reduce the property value—and thus the property tax bill—by limiting the future development of the property with a conservation easement. Property tax levels are measured with a variable equal to the real per capita property revenue of each state in the first year of each period.

Hypothesis 7: Conservation easement donation rates will be greater in states with high per capita incomes than in states with low per capita incomes.

Daugherty (1977) suggests that conservation easements may make financial sense only for wealthy donors, leading one to expect greater donation rates in wealthier states. The analysis includes a variable measuring real per capita income for each state during the first year of each period. Unfortunately, it is likely that there is endogeneity between this income variable and the marginal income tax rate variable making it difficult to disentangle their separate effects in this analysis.

Hypothesis 8: Conservation easement donation rates will be lower in states with high value land than in states with low value land.

Hypothesis 9: Conservation easement donation rates will be lower in states in which land values are increasing at a high rate than in states in which the rate of increase is low.

Where land values are high, conservation easement donation rates are expected to be low because the donor would be forfeiting larger potential revenues from development. If land values are increasing rapidly, land owners may prefer to preserve their ability to develop the land in the future (Coughlin, Berry, and Plaut 1978). Land values are measured as the real average value of farmland in each state for the first year of each period. Land value growth rate is the average annual percentage increase in real farmland values in each state during each donation period.

Hypothesis 10: Conservation easement donation rates will be higher in states that are highly urbanized than in less urbanized states.

Hypothesis 11: Conservation easement donation rates will be higher in states which are experiencing high rates of urbanization than in states in which urbanization rates are low.

Hypothesis 12: Conservation easement donation rates will be higher in more densely populated states than in more sparsely populated states.

Hypothesis 13: Conservation easement donation rates will be higher in states with rapidly growing populations than in states with slowly growing populations.

Hellerstein et al. (2002) find that conservation program participation is greater in areas experiencing urbanization and population growth. Urbanization is measured as the estimated amount of developed land in each state for the year prior to the beginning of each period, expressed as acres per 1,000 acres of total area. The urbanization rate is the average annual percentage increase in the amount of developed land in each state during each period under study. Population density is measured at the beginning of each period, expressed as persons per 1,000 acres. Population growth rate is the average annual percentage increase in population during each period under study.

Hypothesis 14: Conservation easement donation rates will be lower in states with relatively large state park holdings than in states with small holdings.

Privately conserved land, such as conservation easement acreage, and publicly conserved land, such as state park systems, are substitutes. When a state has a large state park system, potential private donors may perceive less need to conserve land through easement donations. State park holdings are measured as total state park acreage per 1,000 acres total area.

Hypothesis 15: Conservation easement donation rates will be greater in states which have experienced relatively large donations in the past than in states that have experienced relatively small donations.

Hypothesis 16: Conservation easement rates will be greater in states in which citizens are highly supportive of conservation policies than in states in which support is low.

State conservation easement donation rates may vary depending on attitudes of potential donors. One indication of citizen attitudes is the proportion of the state’s area that has already been protected by conservation easements. States that have experienced high donation rates in the past can be expected to continue experiencing high donation rates in the future.

States also vary in the extent to which citizens are supportive of conservation policies. Voting records of a state’s congressional delegation are often used in developing measures of citizen ideology (Berry et al. 2010). The League of Conservation Voters (LCV) rates each state’s congressional delegation according to how frequently the members voted in support of LCV-approved policies. The analysis includes a variable measuring the average score for each state’s congressional delegation during each period under study.

Hypothesis 17: Conservation easement donation rates will be greater during later periods than during earlier periods.

Land trust census data indicate that conservation easement donations have increased sharply since the mid-1990s (LTA 2006). To account for this increase, the analysis includes a time trend variable that takes a value of zero for observations occurring during the first time period and increases by one for each succeeding period.

Empirical Results

The model is estimated using both OLS regression and panel regression. For each regression method, two versions of the model are estimated. One version models the mere presence of tax credits; the other version models tax credits using the percentage of fair market value used to compute them. The leftmost columns of Table 1 present the estimated coefficients of the two regressions using OLS.

Table 1.

Ordinary Least Squares and Panel Regression Results Predicting Conservation Easement Donation Rates, 1994–2005

	Ordinary Least Squares				Panel Regression
	Coeff.	SE	Coeff.	SE	Coeff.	SE	Coeff.	SE
Tax credit	0.240	0.161	—	—	−0.459	0.325	—	—
Period prior to first period	0.120	0.134	0.106	0.131	−0.353^†	0.195	−0.321^†	0.177
Tax credit × First period	0.713^**	0.236	—	—	0.620^*	0.243	—	—
Tax credit level	—	—	0.001	0.004	—	—	−0.003	0.005
Tax credit level × First period	—	—	0.016^***	0.004	—	—	0.018^***	0.004
Transferable tax credit	0.241	0.236	0.466^*	0.199	0.210	0.375	−0.130	0.282
Tax credit cap < $100k	−0.242	0.216	−0.153	0.211	0.171	0.352	−0.169	0.310
Time	−0.009	0.040	−0.001	0.041	−0.102	0.105	−0.104	0.103
High-income marginal income tax rate (%)	0.020^†	0.011	0.025^*	0.011	−0.021	0.031	−0.013	0.032
Real per capita property tax revenue ($1,000)	−0.074	0.147	−0.018	0.154	−0.165	0.554	−0.151	0.538
Farm land value ($1,000/acre)	−0.129^*	0.059	−0.103^†	0.058	−0.322^**	0.119	−0.303^*	0.117
Farm land value annual growth rate (%)	0.002	0.008	0.005	0.008	0.000	0.010	0.008	0.011
Developed land (acres/1,000 acres total area)	−0.002	0.002	−0.001	0.002	0.007	0.012	0.009	0.011
Developed land annual growth rate (%)	0.011	0.038	0.028	0.038	−0.019	0.056	−0.011	0.055
State parkland (acres/1,000 acres total area)	−0.012^*	0.005	−0.011^*	0.005	−0.033^***	0.010	−0.033^***	0.010
Population density (persons/1,000 acres total area)	0.002^***	0.000	0.001^***	0.000	0.004	0.007	0.002	0.007
Population annual growth rate (%)	−0.024	0.050	−0.038	0.050	−0.077	0.142	−0.179	0.142
Real per capita income ($1,000)	0.031^*	0.014	0.030^*	0.015	0.082^†	0.042	0.074^†	0.041
LCV score	−0.001	0.001	−0.001	0.001	0.000	0.004	0.001	0.004
Prior donations (acres/1,000 acres total)	0.095^***	0.010	0.088^***	0.010	0.031	0.023	0.034	0.024
Constant	−1.329^*	0.599	−1.628^**	0.568	−1.139	1.174	−1.100	1.177
	N = 171		N = 171		N = 171		N = 171
	Adj. R ² = .63		Adj. R ² = .64		Adj. R ² = .72		Adj. R ² = .73
	F(18, 152) = 16.77		F(18, 152) = 17.64		F(64, 106) = 7.74		F(64, 106) = 8.36
	p < .001		p < .001		p < .001		p < .001

Note: LCV = League of Conservation Voters.

^† p < .1. ^*p < .05. ^**p < .01 (all two-tailed tests). Estimates are unstandardized regression coefficients.

The OLS regression coefficients indicate that several of the predictor variables, including some of the tax credit variables, are statistically significant at the 95 percent confidence level or higher. The results presented in the first column of Table 1 indicate that the presence of a nontransferable, noncapped state tax credit increases the predicted long-run easement donation rate by 0.24 from 0.28 to 0.52 acres per 1,000 acres eligible per year, holding all other predictors at their average values. This effect is substantively significant, as it represents an increase greater than 85 percent. It is not, however, statistically significant. A transferable, noncapped tax credit has a much larger and statistically significant effect, increasing the predicted long-run easement donation rate by 0.48 from 0.28 to 0.76 acres per 1,000 acres per year. The coefficient of the tax credit cap indicator variable has the expected negative sign and although the effect of the cap on the donation rate is substantively large, −0.24 acres per 1,000 acres eligible per year, it is not statistically significant. The impact of a tax credit during the first period of implementation is approximately 1.24 acres per 1,000 acres eligible per year. This impact is not only statistically significant but is significantly greater than the long-run effect of the credit by approximately 0.96 acres per 1,000 acres eligible per year. Contrary to Hypothesis 3, these results indicate that there is an increase in donations during the period prior to implementation of the tax credit, although the coefficient is not statistically significant.

The results presented in column 2 of Table 1 model tax credits according to the percentage of fair market value used in computation rather than their mere presence. These results indicate that a nontransferable, noncapped state tax credit equal to 50 percent of fair market donation value (a typical tax credit value) has a very small and statistically insignificant long-run effect on easement donation rates, increasing the predicted donation rate by 0.04 acres per 1,000 acres eligible per year, holding all other predictors at their average values. The impact during the initial period of implementation is larger than the long-run effect by approximately 0.81 acres per 1,000 acres per year, a statistically significant difference. Like the results in column 1, these results indicate an increase in donations during the period prior to implementation of the tax credit. If the 50 percent tax credit is transferable, the estimated long-run impact is much larger, increasing the predicted donation rate by 0.51 acres per 1,000 acres per year, a statistically significant impact.

The other statistically significant variables indicate that higher tax rates, greater incomes, higher population densities, and larger prior donations are all associated with greater donation rates, as hypothesized. Higher land values and larger state park holdings are, as expected, associated with lower donation rates. The estimated effects of these variables are similar in both version of the OLS model. An increase in the marginal tax rate of 2.5 percentage points (roughly 1 SD) increases the predicted donation rate by 0.06 acres per 1,000 acres per year, which is an increase of more than 20 percent. An increase in per capita income of $4,000 (less than 1 SD) increases the predicted donation rate more than 40 percent, or 0.12 acres per 1,000 acres per year. An increase in population density of 350 persons per 1,000 acres (roughly 1 SD) increases the predicted donation rate by more than 150 percent, or 0.45 acres per 1,000 acres per year. An increase in farm land value of $1,000 per acre (much less than 1 SD) decreases the predicted donation rate by 0.10 acres per 1,000 acres, more than 34 percent. An increase in state park land holdings of 15 acres per 1,000 acres (roughly 1 SD) decreases the predicted donation rate by 0.17 acres per 1,000 acres, or about 58 percent.

By analyzing these data with OLS regression, all the data are pooled together and the analysis does not account for the effects of any unobserved variables that differ between states. Because data are available for multiple time periods, the data can be analyzed using a fixed effects panel regression. This regression includes dummy indicator variables for each state, which allows the analysis to control for the effects of these interstate differences. The results of panel regressions are presented in the rightmost columns of Table 1.

Both sets of panel regression results indicate that tax credits have statistically significant and positive impacts only during the initial period of implementation. The results of the version presented in column 4 indicate that during the initial implementation period a nontransferable, noncapped state tax credit equal to 50 percent of fair market donation value increases the predicted easement donation rate by 0.72 from 0.35 to 1.07 acres per 1,000 acres eligible per year, holding all other predictors at their average values, a magnitude roughly similar to the one estimated by the corresponding OLS model. The results also indicate a substantively large and statistically significant decrease in donation rates during the period prior to implementation, implying that a large portion of the initial increase is a result in a temporal shift in donations, rather than a true increase. The estimated long-run effect of a tax credit is small and negative, although not statistically significant.

While these results indicate that tax credits have no substantively or statistically significant long-run effect after controlling for unobserved differences across states, there are still three predictors with statistically significant impacts, all in the hypothesized direction and all substantively large: incomes, land values, and state park holdings. An increase in per capita income of $4,000 (less than 1 SD) increases the predicted donation rate by 0.29 acres per 1,000 acres per year, an increase of more than 82 percent. An increase in farm land value of $1,000 per acre (much less than 1 SD) decreases the predicted donation rate by 0.30 acres per 1,000 acres, a reduction of about 85 percent. An increase in park land holdings of 15 acres per 1,000 acres (roughly 1 SD) decreases the predicted donation rate by 0.49 acres per 1,000 acres, a reduction of more than 100 percent of the average.

Conclusions

It is easy to see the appeal of conservation easement tax credits during recent decades as policy makers have increasingly relied on market-like mechanisms, rather than command and control regulations, and tax expenditures, rather than direct outlays. The tax credits allow legislators to subsidize conservation, thus gaining the support of environmentalists, in a way that overcomes many of the likely objections of fiscal or social conservatives. They can enact the tax credits and then wait for the market to bring landowners and land trusts together.

The results of this analysis, however, do not make a strong case in support of conservation easement tax credits as a cost-effective alternative for states to conserve undeveloped land and slow sprawling development. None of the models indicated that the tax credits have a statistically significant positive long-run impact on conservation easement donations. The main effect of the tax credits appears to be a sizable and statistically significant “bump” in donations during the initial period of implementation, following a sizable decrease in donations during the prior period. Apparently, these tax expenditures merely subsidize conservation easement donations that would have occurred anyway.

I contend that the findings presented here support a more active role for states in conserving land than the mere offering of tax credits. One of the most robust findings in this analysis is that conservation easement donations are influenced by the extent of a state’s park system. In every model, larger state park holdings are found to be associated with statistically significant and substantively large reductions in conservation easement donation rates, indicating that when states preserve land directly, there is less need to provide conservation privately.

Based on these findings, I argue that states should simply appropriate funds to conserve land directly through the purchase of conservation easements. This method of conservation would likely have greater transaction costs than providing a tax credit, because state officials would now be responsible for identifying appropriate parcels and negotiating purchases, rather than outsourcing this task to landowners and land trusts. On the other hand, by taking responsibility for these decisions, officials would be able to focus state financial resources on preserving land with the greatest conservation value or land that is at most risk of loss to development. Land trusts can, however, still play an important role in the process. In both Maine and Vermont coalitions including land trusts among their membership managed to arrange the preservation of large tracts of land without tax credits. These transactions are a potentially fruitful area for future research to be investigated by means of a detailed case study to identify the factors that lead to the successful negotiation of large conservation transactions.

Footnotes

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

The author received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Alcoba

Natalie

. 2003. Ownership proves key to efforts in U.S.: Conservation groups, states and companies buy acres of land to restrict development. The (Montreal) Gazette (August 6, 2003). A4.

Anderson

John E.

1993. Land development, externalities, and Pigouvian taxes. Journal of Urban Economics 33:1–9.

Berry

David

. 1976. Preservation of open space and the concept of value. American Journal of Economics and Sociology 35:113–24.

Berry

William D.

Fording

Richard C.

Ringquist

Evan J.

Hanson

Russell L.

Klarner

Carl E.

. 2010. Measuring citizen and government ideology in the U.S. states: A re-appraisal. State Politics and Policy Quarterly 10:117–35.

Boyd

James

Caballero

Kathryn

David Simpson

. 1999. The law and economics of habitat conservation: Lessons from an analysis of easement acquisitions. Discussion Paper 99 32, Resources for the Future, Washington, DC. http://www.rff.org/rff/Documents/RFF-DP-99-32.pdf (accessed March 3, 2012).

Brewer

Richard

. 2003. Conservancy: The land trust movement in America. Lebanon, NH: University Press of New England.

Brueckner

Jan K.

2000. Urban sprawl: Diagnosis and remedies. International Regional Science Review 23:160–71.

Business Wire. 1999. Largest conservation partnership in U.S. history completes purchase of 300,000 Acres in New York, Vermont and New Hampshire from champion international corporation. [News Release]. (August 6). Accessed March 3, 2012 from Lexis-Nexis.

Conservation Resource Center. 2007. State conservation tax credits: Impact and analysis. Boulder, CO: Conservation Resource Center. http://www.taxcreditexchange.com/documents/realfinalversion.pdf (accessed March 3, 2012).

10.

Coughlin

Robert E.

Berry

David

Plaut

Thomas

. 1978. Differential assessment of real property as an incentive to open space preservation and farmland retention. National Tax Journal 31:165–79.

11.

Daugherty

Arthur B.

1977. The economics of federal tax incentives for conservation easement donation. National Tax Journal 30:171–82.

12.

Defenders of Wildlife. 2002. Conservation in America: State government incentives for habitat preservation. Albuquerque, NM: Defenders of Wildlife. http://www.defenders.org/resources/publications/programs_and_policy/biodiversity_partners/conservation_in_america.pdf (accessed March 3, 2012).

13.

Defenders of Wildlife. 2006. Incentives for biodiversity conservation. Albuquerque, NM: Defenders of Wildlife. http://www.defenders.org/resources/publications/programs_and_policy/science_and_economics/conservation_economics/incentives_for_biodiversity_conservation.pdf (accessed March 3, 2012).

14.

Feenberg

Daniel

Coutts

Elisabeth

. 1993. An introduction to the TAXSIM model. Journal of Policy Analysis and Management 12:189–94.

15.

Hellerstein

Daniel

Nickerson

Cynthia

Cooper

Joseph

Feather

Peter

Gadsby

Dwight

Mullarkey,

Daniel

Tegene

Abebayehu

Barnard

Charles

. 2002. Farmland protection: The role of public preferences for rural amenities. Agricultural Economic Report No. 815, United States Department of Agriculture, Washington, DC. http://www.ers.usda.gov/publications/aer815/aer815.pdf (accessed March 3, 2012).

16.

Irwin

Elena G.

2002. The effects of open space on residential property values. Land Economics 78:465–80.

17.

Kline

Jeffrey

Wichelns

Dennis

. 1998. Measuring heterogeneous preferences for preserving farmland and open space. Ecological Economics 26:211–24.

18.

LTA. n.d. Land Trusts. http://www.landtrustalliance.org/land-trusts (accessed March 3, 2012).

19.

LTA. 2006. 2005 National Land Trust Census Report. http://www.landtrustalliance.org/land-trusts/copy_of_land-trust-census/2005-national-land-trust-census (accessed October 1, 2010).

20.

National Bureau of Economic Research. n.d. TAXSIM related files at the NBER. http://www.nber.org/∼taxsim/ (accessed March 3, 2012).

21.

New England Forestry Foundation. 2001. The Pingree Forest Partnership. http://www.newenglandforestry.org/conservation/pingree.shtm (accessed September 2, 2010).

22.

Parker

Dominic P

. 2002. Cost-effective strategies for conserving private land: An economic analysis for land trusts and policymakers. Bozeman, MT: PERC. http://www.perc.org/pdf/land_trusts_02.pdf (accessed March 3, 2012).

23.

Simpson

R. David

. 2002. Tax rules, land development, and open space. Discussion Paper 02-61, Resources for the Future, Washington, DC. http://www.rff.org/Publications/Pages/PublicationDetails.aspx?PublicationID=17290 (accessed March 3, 2012).